Vehicle checking method based on video image characteristic

Abstract

The invention provides a vehicle detection method based on video image characteristics. The invention makes use of a panoramic digital camera for collection of monitored video images of a plurality of lanes. The field of view of the panoramic digital camera is pointed at all of lanes in a direction to collect the monitored video images of a plurality of lanes in real time. Digital video images of each frame acquired by the panoramic digital camera are analyzed through a vehicle detection algorithm based on the video image characteristics. The positions and the occurrence time of the vehicles in the images are found out. At the same time, video snapping signal is generated, which drives a high speed CCD industrial camera to snap panoramic photographs. The images of the targets can be cut precisely from panoramic images according to the position information of the targets positioned in video vehicle detection, and are used for identifying number plates, punish vehicles breaking regulations and assist in public security. The invention improves the reliability and accuracy of the video vehicle detection algorithm by comprehensively using the fringe characteristics and the symmetry properties of vehicle figures, and detects vehicles quickly. In addition, the invention can be adapted to exterior environmental changes well.

Description

technical field [0001] The invention relates to the field of intelligent traffic monitoring, in particular to a vehicle detection method based on video image features. Background technique [0002] With the development of the economy and the improvement of people's living standards, cars have played an indispensable role in people's daily life, but at the same time it has brought heavy pressure to urban traffic management. Nowadays, intelligent transportation has become the focus of public attention, and vehicle detection is the key technology and foundation of vehicle monitoring and traffic flow statistics in intelligent transportation, and its detection efficiency directly affects the performance of vehicle capture and license plate detection rate. [0003] At present, the traditional means of vehicle detection is to bury the ground induction coil under the road surface. When the vehicle passes the ground induction coil, the inductance of the ground induction coil changes,...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to overcome the shortcomings of existing vehicle detection methods that are complicated to implement, each lane needs to correspond to a digital camera, redundant equipment, low resolution, and high rate of false shots and missed shots, and provides a video-based A vehicle detection method based on image features, which is simple to implement, highly reliable, and highly adaptable to the environment. Through the cooperation of digital cameras and high-speed CCD industrial cameras, multiple lanes can be monitored at the same time, and high-definition panoramic images and close-ups of vehicle areas can be obtained quickly and efficiently. image

Images